This paper presents an accelerated workflow for nanoparticle structure characterization using artificial datasets. The method incorporates a new shading model based on secondary electron imaging contrast to accurately simulate nanoparticles and develop software for simulating SEM images and automatically generating annotations. The study demonstrates the effectiveness of the proposed method by generating a dataset of Cu2O crystals with six different shapes. The authors utilize the Mask R-CNN model for particle detection and size measurement, achieving satisfactory results. Due to the lack of ground truth annotations, we utilized our developed image segmentation software, allowing researchers to manually correct the segmentation results based on our model's automatic segmentation, with the corrected results used as ground truth annotations. Compared to manual measurement methods, our approach is more convenient, efficient, and can ensure a certain level of accuracy. In practical use, technicians can assess the reliability of size measurements based on the visualized instance segmentation results. Overall, the proposed method provides a cost-effective and efficient alternative to manual annotation and demonstrates promise in various fields.